The long term objectives of this proposal are to develop Acarbose as a chemopreventive agent for colon cancer and to define intermediate markers of colonocyte differentiation in order to monitor the mucosal effects of Acarbose in a chemoprevention trial. With demonstrated chemopreventive activity Acarbose could be used in subjects at risk for colon cancer, reducing colon cancer incidence and the need for invasive monitoring. Sensitive histologic markers of differentiation could identify those at risk and could be used to monitor clinical trials. Acarbose may act as a chemopreventive agent by providing butyric acid prone substrate for colonic fermentation. Ongoing fermentation drives butyric acid absorption by colonocytes. Butyric acid may maintain mutated colonocytes in a normal state until they are sloughed into the colonic lumen. Studies supporting this hypothesis have shown that butyric acid has antineoplastic activity in vitro and in vivo, that butyric acid is a powerful mucosal differentiating agent, and that higher molar percentages of butyric acid are present in colonic contents of normals than those with colonic neoplasia. Cytoskeletal proteins are not fully expressed in some colon cancer cells. Butyrate can induce cytoskeletal protein expression. Therefore the mucosal distribution of cytoskeletal proteins may serve as intermediate markers of mucosal differentiation. These markers may be modulated by colonic butyric acid production. The specific aims based on these hypotheses are: 1) To determine a potentially chemopreventive dose of Acarbose in an 8 week trial (4 week cross over) using fecal butyric acid percent of total short chain fatty acids as an intermediate marker of activity, and to determine the short term effect of Acarbose on fiber polysaccharide excretion. 2) To determine whether intermediate markers of Acarbose activity can be maintained in a 12 month trial (6 month cross over) and whether Acarbose has long term effects on the colonic microbial community and fecal excretion of fiber polysaccharides. 3) To define a combination of histologic markers (cytoskeletal proteins p52 and p35, and bromodeoxyuridine uptake) that can be used as an intermediate differentiation marker in a chemoprevention trial. 4) To determine the effect of Acarbose on histologic markers of colonic mucosal differentiation.